Nobody to Shoot? Power, Structure, and Agency: a Dialogue. Journal

Home , Clarissa Rile Hayward, Structure and agency

This article was downloaded by: On: 30 January 2009 Access details: Access Details: Free Access Publisher Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Power Publication details, including instructions for authors and subscription information: http://www.informaworld.com/smpp/title~content=t778749998

Nobody to shoot? Power, structure, and agency: A dialogue Clarissa Hayward a; Steven Lukes b a Washington University in Saint Louis, Political Science, b New York University, Sociology,

Online Publication Date: 01 April 2008

To cite this Article Hayward, Clarissa and Lukes, Steven(2008)'Nobody to shoot? Power, structure, and agency: A dialogue',Journal of Power,1:1,5 — 20 To link to this Article: DOI: 10.1080/17540290801943364 URL: http://dx.doi.org/10.1080/17540290801943364

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. Journal of Power Vol. 1, No. 1, April 2008, 5–20

Nobody to shoot? Power, structure, and agency: A dialogue Clarissa Haywarda* and Steven Lukesb

aWashington University in Saint Louis, Political Science; bNew York University, Sociology TaylorRPOW_A_294502.sgm10.1080/17540290801943364Journal0000-0000Original20081000000AprilClarissaHaywardchayward@wustl.edu and& of Article Francis Power (print)/0000-0000Francis 2008 (online)In this dialogue, Clarissa Hayward and Steven Lukes debate the relation among power, social structure, and human agency. The authors converge on the view that not only moral responsibility, but also political responsibility is relevant to the study of power. They disagree about how to analyze difficult cases in which some agents are clearly subject to social constraints on freedom, but no powerful actors seem responsible for their constraint. Keywords: agency; determinacy; power; responsibility; structure

Concerns about human agency and about social structure have long been central to the study of power. Beginning with Michel Foucault, contemporary social and political theorists (especially, but not exclusively, theorists who self-identify as postmodernists or poststructuralists) have prob- lematized the relation among these three terms.1 In so doing, they have generated important insights – insights about the normalizing power of constructed social identities, for example – which were not thematized in the postwar power debate in Anglo-American political and social science.2 Yet they have done so while largely neglecting the normative concerns that animated that debate. Indeed, one of the striking oddities about work on power from the middle of the last century to the present day is that these two debates – the latter, which originated with the so-called ‘community power studies’ of the 1950s, and the former, largely inspired by Foucault’s genealogical works – have developed parallel to, while only rarely engaging, one another.3 The failure of these approaches to confront each other head on represents a loss on both sides.

Downloaded At: 02:36 30 January 2009 Participants in the Anglo-American power debate have yet to come to terms with Foucauldian claims about the ways identities, norms, and other social constraints typically thought of as ‘structural’ differentially enable and constrain what human agents can do, and what they can be. At the same time, Foucauldians have yet to articulate a coherent normative account of the grounds on which one might criticize particular power relations, and to elaborate avenues for challenging and changing them.4 In this essay, our aim is to bring these two sets of concerns face to face. We decided to write the essay together, because we share nontrivial premises about how power shapes human freedom, and about how social theorists and social scientists should study it. We both reject strongly determinist positions, which deny the existence and/or the importance of human agency. At the same time, we both reject naively voluntaristic views, which reduce power to the conscious and intentional actions of agents. What is more, we both share the conviction that analyzing power relations is an inherently evaluative and critical enterprise, one to which questions of freedom, domination, and hierarchy are – and should be – central.5 We decided to structure the essay as a dialogue, however, because important differences separate our views of power, structure, and agency, and because our aim is in large part to explore and to clarify (if not resolve) those differences. One of us, Steven Lukes, believes the importance of

*Corresponding author. Email: [email protected]

the notion of responsibility to claims about power makes the latter, necessarily, an agent-centered concept. The powerful, by this view, are those actors (individual or collective) who can reasonably be held responsible for limits imposed on the freedom of other actors. The other, Clarissa Hayward, thinks that because constraints typically identified as ‘structural’ are social in origin, and because their effects are very often amenable to change, they are relevant to the normative project that drives the power debate. That project is badly served, her claim is, when those who study power fail to analyze and to evaluate structural constraints on freedom that are significant, inegalitarian, and remediable. The essay proceeds as follows. In the first and second sections, we present, in abbreviated form, our arguments – first articulated in PRV and DFP – about the relation among power, social structure, and human agency. In the third and fourth, we further focus our disagreement, respond- ing to each other’s principal claims. The exchange helps to clarify the difference between our views, and it brings to light important points of agreement. Perhaps most significant among the latter is our shared belief that multiple forms of responsibility are relevant to the analysis of power relations: not just moral responsibility, understood as liability, or worthiness of blame (or of praise), but also political responsibility for those outcomes to which multiple agents contribute, without any necessarily being ‘at fault.’ The former – our differences – center on how to grapple with those cases in which some agents seem clearly to be subject to power, but no identifiable actors seem responsible for their constraint. Should such cases be analyzed, as Hayward has suggested, using the language of power and of domination, or are the latter terms, as Lukes has argued, necessarily ‘agency’ terms?

1. Power and agency (Steven Lukes) In his fine book Power: A Philosophical Analysis, Peter Morriss asks an excellent and insuffi- ciently addressed question: ‘Why do we need concepts of power?’ (2002, ch.6) What jobs, in other words, do we want concepts of power to do? What is our purpose in studying power? Morriss’ answer is threefold. We want to know, in practical contexts, what our capacities and the capacities of others are, in order to achieve desired outcomes. (As Brian Barry has remarked, the

Downloaded At: 02:36 30 January 2009 CIA needs to know whom to bribe). We want to know, in moral contexts, whom to hold responsible (whom to blame, and I would add, whom to praise) for outcomes that affect the interests of others. Moreover, we want to know, in evaluative contexts, when we are judging social systems, to what extent they give their citizens freedom from the power of others, and to what extent citizens have the power to meet their own needs or wants. Morriss’ question and his answers relate to power in the general sense of ‘the capacity to bring about effects.’ It is a useful exercise. His answers, however, as we will see, need some refinement and extension. In my book, PRV, I focus, more narrowly than does Morriss, on power over others and, more narrowly still, on power over others that is dominating, and yet more narrowly still, on what I call the third dimension of such power, where the power consists, not in prevailing over the opposition of others, nor in imposing an agenda on them, but in influencing their desires, beliefs and judgments in ways that work against their interests.6 This third dimension of power is usually hidden from direct observation; it has to be inferred via the postulation of relevant counterfactuals, to the effect that but for the exercise of the power in question those subject to it would have thought and acted otherwise, in accordance with their ‘real’ interests. Considering this third dimension of power, raises two sets of difficult questions. The first concerns the ‘subjects’ of power: that is, those actors or agents who are subject to it. The central set of questions here centers on how to justify the ascription of real interests to such subjects. The second set of questions raised by power’s third dimension concerns the ‘sources’ of power: that is, those agents or actors who have or who exercise power. Here the central set of questions centers Journal of Power 7

on whether power is to be conceived as located in agents, individual or collective, or, alternatively, in impersonal structures. In the Anglo-American debate, from which PRV stemmed, the question of real interests was highly controversial. The very idea of real interests was denounced as un-, even anti-empirical, or arrogantly paternalist, or crypto-Marxist.7 Questions centered on the subject of power raised, in that debate, further questions about the normative grounds for criticizing domination. In the ‘continental’ manner of discussing power that stems from Foucault, by contrast, these normative questions have not been an issue; perhaps because the answer to them was presumed to be obvious. Instead, controversy has centered on issues of agency and structure. Taking up Morriss’ question, my (Anglo-American) view is that we need an agent-centered notion of dominating power in three dimensions. Agency must, however, be conceived broadly rather than narrowly. A narrow agent-centered view of power entails that power is only at work when it is exercised intentionally and positively. On a broad view, by contrast, agents can have power that they never exercise, and they can have power the effects of which they do not intend. What is more, these effects may result from inaction, or failure to act. The central reason for insisting on this broad view of agency, to which I shall come, is that we can and often do hold agents responsible for consequences they neither intend nor positively intervene to bring about. We need such a view of power for all three of Morriss’ purposes, and more. In practical contexts, we need to be able to locate and track the agents, individual and collective, who have the capacity to affect our interests, as well as those of others. If we are liberals, our practical concerns will center on how to devise and maintain institutional structures that restrain the powerful from harming others’ interests and from limiting their freedoms. Such restraint prevents the powerful from monopolizing first-dimensional decision- and policy-making power, and limits their second-dimensional attempts to control agendas. It also limits their third-dimensional capacity to frame public issues in a way that distorts or suppresses people’s perceptions of their interests. In moral contexts, we need to know whom to hold responsible for outcomes that seriously affect the interests of others, especially where those outcomes are harmful and their effects wide- spread. Consider the case, raised in PRV, of the lack of access to decent and affordable housing by the poor and racially oppressed. Insofar as this problem is due to the action or inaction of iden-

Downloaded At: 02:36 30 January 2009 tifiable individuals or groups or institutions, that, by acting otherwise, could have made a difference, then it makes sense to see the latter as powerful because responsible. The most obvious (first-dimensional) example is that of landlords who discriminate against would-be renters. However, other sources of power can lead to this outcome: municipal officials who bend or break the law, and other officials, who, even if not corrupt, hold positions and take routine decisions and such that they could, in principle, make a difference. They are powerful to the extent that they fail to address remediable problems: problems they may not even see as problems. The housing patterns that obtain in older American cities are the effects of power to the extent that they are, in these ways, caused by the action and inaction of such powerful agents. What this case shows is that we have a need to assign responsibilities, and that doing so involves identifying the powerful. It shows that there is a link between power and responsibility: that part of the point of locating power is to fix responsibility for consequences held to flow from the action, or indeed inaction, of specifiable agents. Often, of course, it will be very difficult, and sometimes it will be impossible, to trace the contribution of particular agents. However, to attribute power – to locate, compare, even try to measure its extent – is, in part, to fix responsibility for outcomes. Faced with the complexities of causation – of intertwined causal chains that stretch from background pre-conditions to immediate triggers – we disaggregate by telling a relevant story, fixing on certain links in the chains based on some criterion of relevance. One such criterion is moral. We seek to identify morally responsible agents, whom we may praise or blame. However, the case in question here – the case of discriminatory housing patterns 8 C. Hayward and S. Lukes

– suggests one respect in which Morriss’ three-fold scheme needs elaboration. For the issue here is not just moral; it is also political. To identify the powerful agents who are responsible for the lack of decent housing for the disadvantaged, to the extent that they are, is to hold them politically accountable: in other words, to view them in a political context. In general, in moral contexts, we tend to employ a backward-looking notion of responsibility, whereas in political contexts we tend to be both backward- and forward-looking: we are concerned with both punishing and rewarding past actions, and with making a difference in the future. In evaluative contexts, we aim to assess societies and their institutions with respect to justice and freedom. Here, as in practical and moral contexts, we need the distinction between structural arrangements and the power (as distinct from the powers) of agents. Social and institutional structures consist in constraints and opportunities, and in roles and norms that limit, guide, and shape individuals’ behavior. These are the frameworks within which politics is conducted and economic and social life functions. Nevertheless, these constraints are never uniquely determining. Agents may or may not seize opportunities. The requirements of roles and norms are always subject to agents’ interpretations. Moreover, the power of agents to achieve their goals and to dominate others is always subject to the impact of unintended consequences. When specifiable individual or collective actors, separately or together, by acting or not acting, can make a difference (and thus be held accountable), it becomes appropriate to use the language of power.

2. Power and structure (Clarissa Hayward) The key to Steven Lukes’ defense of an agent-centric conceptualization of power is his claim – first articulated in PRV, and re-stated in section one, above – that ‘the point … of locating power is to fix responsibility for consequences held to flow from the action, or inaction, of certain specifiable agents.’ (58). This claim is, in many ways, intuitive. As Lukes notes, when faced with a case of significant and avoidable constraints on human freedom, such as those we find in many older cities in the contemporary United States, a common impulse is to search for the actors responsible for those constraints, and identify them as ‘powerful.’ Even Lord Acton, after all, following his famous dictum: ‘Power tends to corrupt, and absolute power corrupts absolutely,’

Downloaded At: 02:36 30 January 2009 added, ‘Great men are almost always bad men’ (Dalberg-Acton 1949, p. 364). Nonetheless, as I argue in DFP, this impulse should be resisted by students of power concerned to analyze, evaluate, and change relations of domination. To explain why, in this second section of our essay, I want to explore in greater depth Lukes’ example, which illustrates at once the appeal and the limits of his strictly agentive view.8 Housing markets in older American cities, Lukes rightly notes, advantage the wealthy and the racially privileged, while they disadvantage the poor, as well as members of racially oppressed groups. One should think about this problem as a structural problem, he suggested when he intro- duced the example in PRV, ‘insofar as it is the uncoordinated and unintended outcome of the independent actions of large numbers of actors pursuing their varied respective interests: renters, home-buyers, mortgage lenders, real-estate planners, and so on.’ ‘But insofar as individuals or groups lack access,’ he continued, ‘because of the actions or inactions of other identifiable individuals or groups, who by acting otherwise could [have made] a difference, then it makes sense to see the latter as powerful because responsible.’9 Lukes offers as examples of such ‘powerful because responsible’ agents, landlords who engage in racial and other types of discrimination; corrupt municipal officials; and politicians at various levels of government, who are positioned such that they could, in principle, ‘make a difference’ (Ibid.). Such agents are powerful, by his view, ‘to the extent that they fail to address remediable problems.’10 The housing patterns that obtain in older American cities are the effects of power’s exercise, if and only if they are caused by the action or inaction of such powerful agents. Journal of Power 9

The crucial distinction Lukes draws here, as elsewhere in PRV, is between limits to what some people can do or can be that are caused by the autonomous actions of others (by ‘the powerful’) and constraints that are not caused by the choices and actions of powerful agents: structural constraints. The difference, he underscores both throughout PRV and above, is not that some forms of constraint are caused by positive action, while others are caused by inaction. A municipal official might exercise power over her poor and working-class constituents, his view suggests, by failing to act when their interests are threatened by a third party, such as a neighboring municipality or a firm. Nor is it that some are foreseen and/or intended, while others are not. An inept official might exercise power over her constituents in ways that she does not consciously intend. Instead, the difference is that some are caused by agents who are able to act in ways that predictably and significantly affect other agents, while some are the unplanned net effect of the actions of multiple actors who could not – not through their individual choices, not through their coordinated efforts – control and direct the outcomes that, together, their actions produce. The former are powerful, because responsible, says Lukes. The latter are not.11 I share with Steven Lukes the conviction that among the most important uses to which the concept of power is put are critical and evaluative uses. Should housing markets advantage the wealthy and the racially privileged, and disadvantage the poor and the racially oppressed? What would be better ways to organize the political economy of the American metropolis? These are the types of questions those who study power typically ask. When we debate about power, that is to say, we do so not only and not principally because we are concerned about how accurately to define a concept, but also, and importantly, because we are concerned to criticize particular relations of power and to identify and evaluate alternatives. The power debate, after all, is a debate driven by a commitment to human freedom and political equality: to the idea that people should have a hand, and that they should have a roughly equal hand, in helping shape the terms that govern their existence. This normative project of criticizing power relations, and identifying and evaluating alternatives, is badly served by an approach that excludes from analysis a priori some set(s) of significant and inegalitarian social constraints. Consider, once more, Lukes’ urban example. His focus is one subset of inegalitarian social constraints: those caused by the action or inaction of agents

Downloaded At: 02:36 30 January 2009 who, in principle, might have ‘[made] a difference’: for example, public officials who engage in exclusionary zoning, perhaps by prohibiting in affluent suburban municipalities the construction of multi-family housing units. Such officials are powerful-because-responsible agents, by Lukes’ view, and the effects of their actions are power effects. Housing patterns in American cities are not power effects, by contrast, if produced unwit- tingly by ‘multiple actors pursuing their varied respective interests.’ Presumably, Lukes would include in this category the processes signaled by the familiar phrases ‘urban economic restructuring’ and ‘white flight.’ Over the course of the last century, as is well known, industrial production in the country was not only automated; it was decentralized at the national level, as well, and, internationally, globalized. Together with technological advances in telecommunica- tions and travel, deindustrialization helped to create an urban economic structure in which multiple individual and collective actors pursuing their varied respective interests – interests in protecting their investments, interests in obtaining an optimal package of public services while paying relatively low taxes – migrated to the suburbs of older cities and to the newer cities of the American West and Southwest. Neither deindustrialization, nor the exodus of middle-class whites from older central cities, was a consciously coordinated process. Their combined effects on those ‘left behind’ were largely unintended. However, together, they eroded the urban tax base at just that historical juncture when African-Americans migrated en masse from the rural South to the urban North, increasing dramatically demand for affordable housing in older cities (Self and Sugrue 2001). 10 C. Hayward and S. Lukes

The example highlights what is intuitive in Lukes’ view. Most people would hold a discriminating landlord morally responsible for the adverse effects of her actions on the housing choices of others, but most would not hold responsible for her unintended role in producing such effects someone who moves her family out of a city in order to live in its suburbs. At the same time, the example illustrates the limits to his agentic-centric understanding of power. Those constraints on human freedom that are typically identified as ‘structural’ are social in origin. What is more, their effects are very often amenable to change. Hence, they are relevant to the critical project that lies at the heart of the power debate: the project of identifying, evaluating, and elaborating methods for changing differential remediable social constraint on freedom. The adjectives ‘social’ and ‘remediable’ in the last sentence are worth underscoring: Neither urban deindustrialization nor the exodus of middle-class whites from older urban centers was the product of inexorable technological and market forces. To the contrary, both were enabled and encouraged by a wide range of human choices and human actions. These include systematic disinvestment in American cities during the postwar years. They include the channeling of resources – both public and private – to new suburbs and to the so-called ‘Sunbelt Cities.’ The differentially constraining effects of these processes were – and they are – amenable to change. Hence, if the structural context of the American metropolis today is one in which some ‘actors pursuing their varied respective interests’ significantly and adversely affect what others can do – their ability to live in safe and decent homes, for instance, their ability to send their children to safe and decent schools – then students of power should analyze the institutions and the practices that produce this context. We should attend to the devolution of governance powers to municipalities, to cite one nontrivial example: powers to make autonomous land use decisions, for instance, and to raise and spend tax dollars on services for residents only. Such ‘local control’ enables and encourages privileged individual and corporate actors to define their interests in terms of removing important collective problems beyond the boundaries of the municipality in which their property lies. Work by legal theorists exploring institutional challenges to local municipal autonomy provides one example of how to criticize and think about changing differential social constraint on freedom, without identifying agents who are morally responsible for that constraint.12 I do not

Downloaded At: 02:36 30 January 2009 mean to suggest that there is anything wrong with holding individual and collective actors morally responsible. However, it would be a mistake to conflate that effort with the project of identifying significant, inegalitarian, and remediable social constraints on human freedom. The types and the forms of social constraint on freedom that (rightly) worry theorists of power are not always caused by agents who are (to recall Lord Acton’s words) ‘bad men.’ What is more, even the ‘bad’ actions of ‘bad’ men often are shaped by remediable social constrains on freedom, which do – and which should – concern theorists of power. One might want to hold morally responsible the discriminating landlord, for example, and the local political official who makes exclusionary land use decisions. At the same time, if the concern is to criticize and to change the overall pattern of such actions and such decisions, one should attend, as well, to the social constraints that structure each agent’s actions. The official, for instance, is significantly constrained by the political control the local demos exercises – a demos from which those who, in order to live in the municipality in question, would need to rent apartments, are effectively excluded. She is constrained by the imperative to compete with other, neighboring municipalities both for mobile capital and for affluent residents. She is constrained by the very interests (interests in protecting property investments, interests in obtaining an optimal package of public services while paying relatively low taxes) that political fragmentation in America’s cities has produced. Even when examining that subset of power-effects produced by agents who seem, intuitively, powerful-because-responsible, if the aim is not to wag a moralizing finger, but to criticize and inform efforts to change relations of power and domination, then it is necessary Journal of Power 11

to examine the structural constraints that help shape those relations. Such constraints often are not produced, through a simple causal chain, by the action or the inaction of readily identifiable ‘bad men.’ Still, as Lukes’ comments in section one suggest, a structural understanding of power does not obscure the relation between power and responsibility. Instead, it directs us to attend, not only to moral, but also to political responsibility. Iris Young has argued persuasively for supplement- ing conventional notions of responsibility-as-liability with an understanding of shared political responsibility for outcomes to which multiple agents contribute, even if in ways that cannot be disaggregated and traced precisely. If the principal purpose of identifying the morally responsible is to assign and distribute praise and blame, the principal purpose of identifying the politically responsible is to enable critique and to motivate change. It is, in other words, the evaluative purpose we pursue when we study power. Even when those who share political responsibility for creating or perpetuating relations of domination are not at fault in a moral sense, their political responsibility obliges them to attempt to understand, and to act collectively to change, the processes and institutions through which, together, they exercise power (Young 2004).

3. The power of the powerful (Steven Lukes) Clarissa Hayward correctly attributes to me the view that the powerful are those whom we judge or can hold to be responsible for significant outcomes. Her analysis developed here and elsewhere, makes me see that this view is more complex than I previously realized. The example of the impact of housing markets brings out an important aspect of the attribution of responsibility that bears on how we locate the powerful and assess the extent of their power – an aspect that is particularly pertinent to political contexts. Unlike Hayward, however, I continue to think that the concept of power should remain attached to the agency that operates within and upon structures. Consider, again, the institutional and other frameworks that shape the constraints and opportunities individual and collective agents face. It is not my contention that these frameworks never contribute to domination. Nor is it my contention that agency and structure are always equally pertinent to what we are trying to explain. In the case of urban restructuring and white flight lead-

Downloaded At: 02:36 30 January 2009 ing to disadvantageous housing for the poor and black, the contribution of political, bureaucratic, and corporate actors may be minor or minimal. However, to the extent that such actors could real- istically make a significant difference to the availability of housing for the poor and black (either by acting separately or in alliance with one another) – whether by challenging exclusionary zoning decisions, say, or by extending the jurisdictional reach of decision-making bodies – then, to that extent, the most appropriate explanation will be in terms of power. Nevertheless, to the extent that white flight and urban deindustrialization are the uncoordinated outcomes of multiple actors pursuing their varied respective interests, with consequent effects on the housing available to the poor and the black, then, to that extent, the explanation must indeed be structural. As I suggested in section one, we attribute responsibility to agents, individual or collective, in several different ways, in different contexts. When we hold them morally responsible, we praise or blame them. When we hold them politically responsible, we call them to account for their decisions and actions, or for their failures to decide and to act. Whether they are legally responsible for some event or situation, such as the bankruptcy of a company, say, depends on the interpreta- tion of the relevant laws. We also sometimes speak of historical responsibility, when we trace some significant outcome back to the actions, and sometimes inaction, of some identifiable actor or actors: for example, the rise of Nazism to the Treaty makers at Versailles, or the Palestinian intifada to the failed summit at Camp David. Discussions of historical responsibility revolve around the question of what difference individuals make to history (Plekhanov 2003, Berlin 1969, Carr 1961). 12 C. Hayward and S. Lukes

In both the moral and the historical cases, when we view an agent as responsible, we typically claim that that agent is causally responsible for the outcome in question. Interestingly, this is not necessarily the case in the legal and political cases. You can be legally responsible for outcomes, such as bankruptcy, which you had no part in causing, by virtue of your official role. In political contexts, you can be called to account to solve problems and to rectify injus- tices to which you made no causal contribution, by virtue of your office. Hayward is, of course, correct in asserting that ‘significant, inegalitarian and remediable social constraints on human freedom’ are not always caused by what Lord Acton called the ‘bad men’ in power. These constraints, indeed, as in the case of white flight to the suburbs, may be best thought of structurally. Those who are in a position to contribute to remedying them, perhaps by working to change the institutional framework, but who did not contribute to causing them, are, to that extent, responsible and thus powerful. Of course, Hayward is also right to argue that they – politicians and officials, for example – will in turn be subject to significant constraints, such as the imperatives of competition and the interests they are elected or appointed to represent. However, that simply shows that their power as agents (in this case to remedy injustice and unfreedom) is limited by the structures – the constraints and opportunities, roles and norms – within which they operate. Hayward’s central claim is that an agent-based conceptualization of power draws attention away from these structures: that it obstructs (in her words, above) ‘the project of identifying significant, inegalitarian, and remediable social constraints on freedom’ and the democratic political project of expanding ‘the scope of the social capacity for people to act in ways that affect the relations and processes in which they participate,’ and overcoming their ‘exclusion … from processes of shaping social constraints on their action’ (Hayward 2006). I fail to see why this should be so. For her own argument amounts to an appeal to collective agency (‘the social capacity for people….’) as a form of democratizing power operating within and upon structural constraints. To which I would add that there are other variants of this idea of the power of those subject to structural constraints to contest and even sometimes transform them. I am thinking of Vaclav Havel’s analysis of the ‘power of the powerless’, in which separate individuals ‘living in truth’ can, by rejecting the symbols of the official power structure, expose and demoralize its

Downloaded At: 02:36 30 January 2009 agents and leaders (Havel et al. 1985). Another example is James Scott’s accounts of the ‘weapons of the weak’ in which small acts of resistance, coded and often hidden, over time, can be contagious and corrosive of dominant structures of power (Scott 1985). Therefore, I do not see why confining the attribution of power to agents obstructs the projects Hayward endorses, and similar such projects. The impersonal, socially generated, and remediable constraints that limit people’s freedoms and opportunities and shape their interests do not have to be identified as sources of power in order to be objects of attention and contestation. Nevertheless, there is a further reason why I believe that retaining the link between power, agency, and responsibility is important. It enables us to keep in focus the very question of the difference that agents can make to outcomes and to cast a critical eye on attempts by powerful agents to escape their own responsibilities by ‘blaming the system.’ Human agents, whether individuals or collectivities, have power or are powerful within structural limits, which enable and constrain their power. The natural way to distinguish between power and structure is to say that we attribute power to agents when it is in their power to act or not to act. They have two-way powers: they have power when it is in their power to act otherwise. If they are so structurally constrained or determined that they are unable to act otherwise than they do, then they are powerless to do so, and so they are powerless, not powerful. They simply enact or transmit the dictates of the structures that uniquely constrain them. However, determining when this is so and when it is not is not a straightforward matter of fact but of judgment (the judgment of both actors and observers) and so highly contentious. Journal of Power 13

To illustrate the importance of retaining the distinction, contrast the case of the housing market we have discussed with another in which it is plainly appropriate to explain the unavail- ability of decent affordable housing to the poor and black in terms of power. At the time of writ- ing, two years after Hurricane Katrina, New Orleans is still suffering from an acute shortage of housing that has nearly doubled the cost of rental units in the city. Yet, as a report in the New York Times comments, ‘official attention to help revive the shattered rental and home apartment market has been scant’ (Saulny 2007). The Federal Emergency Management Agency (FEMA) has announced the closure of all the trailer camps it runs for the hurricane’s victims, and Ronnie Simpson, a FEMA spokesman, says: ‘It’s the individual’s responsibility to go out and find what’s suitable for them’ (Ibid.). Meanwhile, the city housing authority has approved the demolition of 4000 public housing units damaged by the storm and will build instead mixed-income projects, large parts of which will not be affordable to previous residents. Martha J. Kegel, executive direc- tor of Unity of Greater New Orleans, a group of 60 agencies that house and feed the homeless, offers a succinct power-and-responsibility analysis of the situation: It will take a long time for replacement affordable housing to be built. To withdraw housing assis- tance to the neediest people is a shirking of federal responsibility for the design failure of the federal levees in New Orleans, which was the cause of most of the destruction of affordable housing here. (Saulny 2007) Compare Kegel’s assessment of who is responsible with the ways some of the central actors in the course of the hurricane sought to gain acceptance of alternative framings of what occurred. According to an earlier New York Times article, those in authority at the federal level were preoc- cupied, both during and immediately after Katrina, with the constitutional constraints within which they took themselves to be placed. Pentagon officials, for instance, hesitated at the prospect of sending combat troops to New Orleans on the grounds that, in the words of Defense Secretary Donald Rumsfeld. ‘The way it’s arranged under our Constitution, state and local officials are the first responders.’ Assistant Secretary of Defense Paul McHale expressed a similar belief when he asked, ‘Could we have physically moved combat forces into an American city, without the gover- nor’s consent, for purposes of using those forces – untrained at that point in law enforcement – for law enforcement duties? Yes.’ However, he continued, ‘Would you have wanted that on your

Downloaded At: 02:36 30 January 2009 conscience?’ (Lipton et al. 2005). Of course, factual questions are in part at issue. The request for federal help was, as the New York Times puts it, ‘ill-defined.’ Before the waters breached the levees, submerging the city, Governor Blanco had told the president, ‘I need everything you’ve got. I am going to need all the help you can send me.’ Yet at a meeting with the President aboard Air Force One, the governor had expressed reluctance to give up her authority. According to the spokesman for the Department of Homeland Security, ‘There was a significant amount of discussions between the parties and likely some confusion about what was requested and what was needed’ (Ibid.). So where did power lie? As the waters rose, did not those in strategic positions have an over- whelming responsibility, that was both moral and political, to bypass formal legal rules? One factor that appears to have prevented decisive and timely federal action was a combination of beliefs in the actors’ minds: belief in the legal impropriety of military domestic intervention in the affairs of states, combined with a general ideological predisposition against centralized public administration. Apparently, these beliefs were held strongly enough to withstand the sense that what was occurring was a catastrophe involving the submerging of an entire city and the displace- ment and deaths of large numbers of citizens. It is noteworthy that not everyone in authority felt so restrained. For instance, Sheriff Warren Evans of Wayne County, Michigan, ignored his gover- nor’s plea to wait for formal requests to send food, water, medical supplies, and a search and rescue team. ‘I could look at CNN and see people dying,’ he said, ‘and I couldn’t in good conscience wait for a coordinated response’ (Ibid.). 14 C. Hayward and S. Lukes

In fact, President Bush did subsequently acknowledge his and, by implication, his administration’s, political responsibility for what had occurred in a televised speech to the nation from New Orleans. He said: Four years after the frightening experience of September 11th, Americans have every right to expect a more effective response in a time of emergency. When the federal government fails to meet such an obligation, I as President am responsible for the problem, and for the solution. (International Herald Tribune 16 September 2005) However, we may ask what this acknowledgment was an acknowledgement of. The President did not provide an account of the foreseeable results of his actions, including his and his officials’ inaction, in the critical days of the hurricane and its immediate aftermath, and he has not done so since. The responsibility he acknowledged ignored the past and, insofar as it looked forward, it was severely limited in scope. It is true that he said that he was prepared to undertake ‘one of the largest reconstruction efforts the world has ever seen,’ but this has turned out to be a hollow promise. It is also true that, in the face of the stark obviousness of the fact that those left homeless and endangered by Katrina were mostly poor and black, the President also acknowledged that their condition had its ‘roots in a history of racial discrimination, which cuts off generations from the opportunity of America.’ Nevertheless, the responsibility acknowledged for the reconstruction and for the far wider issues of poverty and inequality has amounted to little indeed. The public debate over the hurricane and its aftermath ranged over these questions of responsibility. Some, such as Rumsfeld and McHale, claimed to be powerless to act because of constitutional constraints. Others, notably the President himself, while admitting responsibility, sought to shift attention from the past to the future. Others countered criticism by denouncing what they call ‘the blame game’ (while encouraging the blaming of Democrats and state and local officials). And even some Democrats, notably Joseph Lieberman, joined in the practice of denigrating the very idea of identifying past responsibilities, suggesting that the problems were not failures of power but of structure, not faults of leadership but of ‘the system’, for which everyone and no- one is responsible: that the disaster was, as the cover of Time (but not the accounts inside) put it, a ‘system failure.’ (Time 19 September 2005) Downloaded At: 02:36 30 January 2009 The point of re-telling the story of this contest among the central actors of how to frame Katrina is twofold. First, it shows how contentious is the issue of where the structural boundaries to power lie: that it is not a straightforwardly factual issue, but involves controversial beliefs and judgments, about which both agents and observers will disagree. Second, it suggests the value of continuing to conceive of power as tied to agency acting within and upon structures. For doing so allows us to keep alive the crucial question of the moral and the political responsibility of the powerful for what they do and what they fail to do in the past and in the future.

4. Power without the powerful (Clarissa Hayward) When agents act, they act within limits that are set, in part, by the actions of other agents. At the same time, they act in contexts that are structured by rules and laws and norms: social boundaries to action, which – not unlike the actions of other agents – limit what they can do and what they can be. As agents act and interact within structural limits, they develop expectations about what it is that one does, and what it is that one ought to do, in particular contexts. They develop not just subjective, but also intersubjective, understandings of the meanings particular actions hold. These social expectations and social meanings always mediate between, on the one hand, structural constraints (such as laws, policies, rules, or norms), and on the other hand, the action and the inaction of human agents. Structure does not determine action. Instead, it shapes action, by rendering some forms, in some contexts, costly or otherwise difficult, while rewarding or Journal of Power 15

otherwise encouraging others. Structure shapes social action through social meanings, which agents continually interpret and re-interpret. This claim is, roughly, what I take Lukes to assume when he writes that, in the case of the Louisiana hurricane, one cause of the early inaction of federal agents was ‘a combination of beliefs in [their] minds.’ It is, of course, a long way from here to the extreme position – a position associated at times (perhaps unfairly) with symbolic interactionism, and with the constructivist movement in the study of international politics – that ‘structures are what agents make of them.’ I do not believe Steven Lukes thinks the latter claim correct. I do not, myself, think it correct. For present purposes, however, it is worth spelling out why it is that structures are not only, or are not simply, what the agents whose fields of action they shape take them to be. Imagine a position in a power relation that is defined by some set of rules or laws or norms. You might think of a social role, such as ‘student’ or ‘teacher’ or ‘parent,’ or an ascribed social identity, such as ‘man’ or ‘woman’ or ‘African-American.’ If, by the norms that define this particular power relation, those who occupy this particular position do or should behave in way x, or if, by those terms, they do or they should exhibit characteristic y, then the other actors who participate in the relation will tend to treat those agents in ways informed by the relevant expectations. Other actors may expect students to work hard and to be deferential, for example. They may expect men to act aggressively and to be dominant in professional and social settings. They may expect women to focus their attention and their energies on domestic life, and to be kind, warm, communicative, and noncompetitive. Of course, if students, or if men or women, face these (or other) expectations, they need not necessarily meet them. Students can be indolent and disrespectful. Men can exhibit passivity, and women, combativeness. However, to the extent that a given student, or a given man or woman, wants to avoid the sanctions attached to nonconformity, to the extent that (s)he wants to obtain the rewards (like good grades and good jobs and status and respect) that a conforming perfor- mance earns – to that extent, (s)he will stick to the script. Structures are more than just meanings and expectations. They are relatively durable meanings and expectations, sustained by systems of reward and sanction, which make some forms of action, if not impossible, then highly improb-

Downloaded At: 02:36 30 January 2009 able, and others, if not inevitable, then exceedingly likely. Structures do not determine action, then, but they produce predictable patterns of action: patterns, which (rightly) capture the attention of those who analyze and who evaluate relations of power. If a defender of the status quo in urban America were to say to Steven Lukes, or if she were to say to me, ‘Some residents of the so-called “inner city” are able to find adequate and affordable housing, which means that those agents are not “powerless,” after all’ – or if, by the same token, a defender of the Bush administration’s handling of Katrina were to suggest to Lukes or to me, ‘Some displaced individuals have found “what’s suitable for them,” which means that in New Orleans, no domination obtains’ – we both would remain entirely unconvinced. Our worry is not that it is impossible for anyone to find decent and affordable places to live in New Orleans, or in America’s central cities. It is that the abilities of some people to do so are significantly constrained; that they are constrained in ways that are social in origin; and that they are constrained in ways they need not be. When those who analyze and evaluate relations of power focus on inter-agentive forms of domination – when we focus on discriminating landlords, for example, who limit housing options for the poor and the racialized – we do not claim the dominant determine the actions of those they oppress. We do not claim the poor and the racialized, for instance, cannot ‘resist,’ de Certeau-like, through ‘tricks,’ by ‘get[ting] away with things’ or ‘maneuver[ing].’ (de Certeau 1984)13 Nor do we claim that the dominated cannot resist through more organized forms of collective action, such as political drives to strengthen anti-discrimination laws. Instead, we claim the actions of the 16 C. Hayward and S. Lukes

dominant significantly limit the fields of possible action of the dominated, and that they do so in ways that are not inevitable, that need not be. Codified and institutionalized human actions – structural constraints – have similar effects. Consider again the example of exclusionary zoning laws. These man-made institutions create patterned asymmetries, which significantly limit the freedom of some human agents, and which limit their freedom in ways that need not be. Those who oppose significant social constraints on freedom should, therefore, study, and we should criticize, these institutions. More generally, those who oppose significant social constraints on freedom should study and should criticize codified and institutionalized human actions that create patterned asymmetries in the social capacity to act. We should do so, I want to underscore, even if and when these constraints work, in whole or in part, by shaping the action of privileged agents. Recall the white middle class homeowners from our urban example. New Deal Era policies such as those of the Home Owners Loan Corporation (HOLC) and the Federal Housing Administration (FHA) – policies which are widely credited with nationalizing the practice of redlining – helped produce what, today, function as differential structural constraints on the actions of the racially privileged and disprivileged. If one thinks back to the time when these policies first were promulgated, the inter-agentive model of power works reasonably well. Those actors who developed and who implemented the FHA mortgage program, for example, significantly and differentially affected the interests of American whites and blacks. They favorably affected the interests of white working- and middle-class families, by dramatically lowering down payments and interest rates, and by extending amortization periods, making it possible, for the very first time, for these actors to purchase and to own private homes. At the same time, they adversely affected the interests of most African-Americans, to whom they denied what would prove, historically, a critical opportunity to accumulate personal wealth.14 However, the FHA mortgage program not only differentially affected interests. It also – by extending loan guar- antees to whites on the condition that they buy homes in racially exclusive enclaves – created a new constellation of racialized interests: interests centered on homeownership, property values, and what George Lipsitz has aptly called ‘the possessive investment in whiteness’ (2006). In 2008, a full four decades after the Civil Rights Movement, and a time when neither the

Downloaded At: 02:36 30 January 2009 FHA nor any other American government agency discriminates in home mortgage lending or insurance, long-standing patterns of public and private investment combine with the political fragmentation of the American metropolis to ensure that ‘various agents pursuing their respective interests’ produce class- and race-based asymmetries in the social capacity to act. These patterned asymmetries are the product of human action. They are not inevitable; they could be challenged and changed. Real challenges to local autonomy along the lines proposed by the legal theorists cited in section two, for example, would go a long way towards disrupting them, as would a return to de facto standards in constitutional review in discrimination litigation.15 These asymmetries should be challenged, what is more, most proponents of non-domination would agree, even if their proximate cause is ‘various agents pursuing their respective interests,’ rather than the choices and actions of agents who fit the description ‘powerful-because-responsible.’ That said, I take Steven Lukes’ point that too much talk along these lines – too much talk of ‘structural constraint’ as the cause of, or an important factor contributing to, domination – can pave the way for evasions of moral responsibility by those who do function as agents of interper- sonal domination. However, it need not. In their capacities as social and political theorists, students of power should concern themselves with patterned asymmetries in the capacity to act, including those that result from structural constraints on freedom. At the same time, in their capacities as social and political actors, they should hold responsible those agents who, through their action or inaction, create significant, predictable, and avoidable constraints on the fields of action of others. Journal of Power 17

If students of power to fail to address in the language of power constraints on freedom that are significant, social, remediable, and structural, what is more, their failure can create problems analogous to those Lukes cites in section three. It can pave the way for evasions of the political responsibility that multiple actors share: responsibility to analyze, to evaluate, and act to change those constraints. In 1956, Charles Tiebout famously defended metropolitan fragmentation as a political structure that maximizes human freedom. Political decentralization and local control, Tiebout claimed, enable multiple governments to offer multiple packages of public services and tax expenditure levels, from which people freely can choose ‘with their feet.’ It is not difficult to identify the normative problems with a model that defines the differential ability to pay for public services as a difference in ‘preference’ (thus implying, implausibly, that those who lack income and wealth prefer not to live in safe and decent neighborhoods, or to send their children to safe and decent schools); that relies on the baldly counter-factual assumption that people are free to move wherever they choose (an assumption which, if fanciful today, was nothing short of incredible in 1956); and that is entirely indiscriminating with respect to the substance of people’s preferences (my preference for Pringles may be morally equivalent to your preference for Doritos, but if I regard my desire to transmit to my child unearned class- and race-based privileges as equivalent to your desire to ensure that your child receives a sound education, I have, to paraphrase Jon Elster (1997), confused the logic of the market with the logic of the forum).16 Still, one can imagine a contemporary Tieboutian asserting, ‘There is nothing wrong with the political fragmentation of the American metropolis, or for that matter, with the place-based inequalities it helps produce. These overall patterns are unintended. They are the result of “various agents pursuing their respective interests.” They are not caused by the choices and actions of morally responsible “bad men.”’ ‘Indeed,’ our naïve Tieboutian might continue, ‘if people want to live in racially and econom- ically inclusive communities they can. After all, structures don’t determine action!’

Conclusion Downloaded At: 02:36 30 January 2009 This dialogue has highlighted not only real differences, but also important commonalities between our understandings of power. We are largely in agreement about what Lukes, in the first section, calls the ‘subjects of power.’ The dominated, the oppressed, the relatively powerless, we both think, are the proper concern of those who study power, because what they can do and what they can be is limited, and significantly so, by social constraints that might be otherwise. Our disagreements, it seems, center on the ‘sources’ of power. Are these only those identifiable agents who are responsible for significant social constraint on the freedom of others? Or should we treat institutionalized human actions (structures) as sources of power, as well? If we think of power, as Lukes urges us, in strictly agent-centric terms, then we may overlook some subset of significant and remediable social constraints on human freedom. Nevertheless, if we think of power, as Hayward suggests, in structural terms, then we may lose sight of those particular agents who are responsible for the constraints we analyze and review. We are reminded of the quandary of the tenant farmer in John Steinbeck’s The Grapes of Wrath, who confronts a tractor driver on the verge of bulldozing his shack. The farmer threatens to shoot the driver, who after all looks to be the (agentive) source of his domination. Nevertheless, the driver strenuously objects: It’s not me. There’s nothing I can do. I’ll lose my job if I don’t do it. And look – suppose you kill me? They’ll just hang you, but long before you’re hung there’ll be another guy on the tractor, and he’ll bump the house down. You’re not killing the right guy. 18 C. Hayward and S. Lukes

‘That’s so,’ the tenant said. “Who gave you orders? I’ll go after him.’ ‘You’re wrong. He got his orders from the bank. The bank told him, “Clear those people out or it’s your job.”’ ‘Well, there’s a president of the bank. There’s a board of directors. I’ll fill up the magazine of the rifle and go into the bank.’ The driver said: ‘Fellow was telling me the bank gets orders from the East. The orders were: “Make the land show profit or we’ll close you up.”’ ‘But where does it stop?’ Steinbeck has his farmer ask the driver of the tractor. ‘Who can we shoot? I don’t aim to starve to death before I kill the man that’s starving me.’ ‘I don’t know,’ the driver replies. ‘Maybe there’s nobody to shoot’ (Steinbeck 1992, pp. 40–41)

Acknowledgements For helpful comments and suggestions, the authors are grateful to Frank Lovett, Ian Shapiro, and the anonymous reviewers for this journal.

Notes 1. See, for example, Connolly (1991), Brown (1995), and Butler (1997). 2. Key contributions to that debate include Dahl (1957 and 1958), Bachrach and Baratz (1962), and Lukes (1974). For an overview and critique, see Hayward (2000). 3. For overviews of the community power studies of mid-century, see Clark (1972) and Ricci (1980). Of Foucault’s genealogical works, the most important for present purposes are Discipline and Punish: The Birth of the Prison and The History of Sexuality, Volume I: An Introduction, but see also Foucault’s more programmatic writings on power, especially (1982) and the essays and interviews collected in Gordon (1980). 4. Fraser (1981), Walzer (1983), Taylor (1984). 5. See Lukes (2005), Power: A Radical View, hereafter PRV, and Hayward (2000), De-Facing Power, hereafter DFP. 6. ‘Prevailing over the opposition of others’ I think of as power’s first dimension, as exemplified by the work of Dahl, Polsby, and others. See, e.g. Dahl (1980) and Polsby (1980). ‘Imposing an agenda on others’ is an example of power’s second dimension, as exemplified by Bachrach and Baratz’s (1962) critique of the first dimensional view. 7. Bradshaw (1976), Polsby (1980); Hay (1997).

Downloaded At: 02:36 30 January 2009 8. My argument in this section draws on my response to Lukes in Clarissa Hayward (2006). 9. Ibid., p. 67, emphasis as in original. 10. Ibid., emphasis as in original. 11. Thus, for Lukes, ‘The powerful are those whom we judge or can hold to be responsible for significant outcomes.’ Ibid, p. 66. 12. See, for example, Ford (1994) and Briffault (1990). 13. James Scott, cited by Lukes (above), draws on de Certeau’s claims about ‘the ingenious ways in which the weak make use of the strong.’ ‘Innumerable ways of playing and foiling the other’s game,’ de Certeau argues, ‘characterize the subtle, stubborn, resistant activity of groups which, since they lack their own space, have to get along in a network of already established forces and representations. People have to make do with what they have. In these combatants’ stratagems, there is a certain art of placing one’s blows, a pleasure in getting around the rules of a constraining space.’ Ibid., xvii, 8. 14. As is well known, the FHA defined racial segregation in the neighborhood surrounding a property as a necessary condition for that property to receive a favorable rating for a federal loan guarantee and specifically recommended the use of racially restrictive covenants to promote segregation. From the start of the agency’s mortgage program through the early 1960s – a period during which the FHA insured mortgages on close to a third of new housing in the US, less than two percent of its state-insured mortgages went to African-Americans, and even these went disproportionately to racially segregated neighborhoods in the South. See Jackson (1987). 15. Under the Warren Court, to trigger a remedy in discrimination cases required demonstrating a pattern of discriminatory effects – or in our terms here, patterned asymmetries in constraints on the action of the subjects of power. The Burger and Rehnquist Courts, by contrast, require showing intention to discriminate on the part of some public agent or agents – i.e. a powerful source of discriminatory effects. For a discussion of the relevant cases, see Shapiro (2003), chapter 3. Journal of Power 19

16. Self-regarding preferences, Elster’s claim is, although often appropriate determinants of individual consumer choices are not, by themselves, appropriate for deciding collective norms: norms which affect, not a sovereign and solitary consumer, but a network of people, who are connected to one another through relations of power.

Notes on contributors Clarissa Rile Hayward, author De-Facing Power (Cambridge University Press, 2000), is Associate Professor of Political Science at Washington University in St Louis. She is currently completing her second book, which focuses on the ways democratic state actors shape political identities through institutions that racialize and privatize urban space. This research has been supported by the National Academy of Education/Spencer Foundation, the National Endowment for the Humanities, and the Institute for Advanced Study in Princeton.

Steven Lukes, author of Power: A radical view (Second edition: Palgrave Macmillan, 2005) is Professor of Sociology at New York University. His previous books include Emile Durkheim: His life and work, Individualism, Marxism and morality, Moral conflict and politics, and Liberals and cannibals: The impli- cations of diversity. His next book is Moral Relativism, to be published by Profile Books in the UK and Picador in the United States.

References Bachrach, P. and Baratz, M., 1962. Two faces of power. American Political Science Review, 56, 947–952. Berlin, I., 1969 [1953]. Historical inevitability. In I. Berlin, ed. Four Essays on Liberty Oxford: Oxford University Press. Bradshaw, A., 1976. A critique of Steven Lukes’ Power: A radical view. Sociology, 10, 121–127. Briffault, R., 1990. Our localism, part ii – localism and legal theory. Columbia Law Review, 90, 346–454. Brown, W., 1995. States of injury: Power and freedom in late modernity. Princeton, NJ: Princeton University Press. Butler, J., 1997 The psychic life of power: Theories in subjection. Stanford, CA: Stanford University Press. Carr, E.H., 1961. What is history? New York: Vintage. Clark, T., 1972. Community power and decision-making. Current Sociology, 20, 6–53. Connolly, W., 1991. Identity/difference: Democratic negotiations of political paradox. Ithaca, NY: Cornell University Press.

Downloaded At: 02:36 30 January 2009 de Certeau, M., 1984., [1980] L’invention du quotidian. Vol. 1 arts de faire. Union Générale d’éditions. In S. Rendall trans. The practice of everyday life. Berkeley: University of California Press. Dahl, R., 1957. The concept of power. Behavioral Science. 2, 201–215. Dahl, R., 1958. A critique of the ruling elite model. American Political Science Review, 52, 463–469. Dahl, R., 1980. The concept of power. In Nelson Polsby, ed. Community power and political theory: A further look at problems of evidence and inference, Second Edition. New Haven, CT: Yale University Press. Dalberg-Acton, J.E.E., 1949 Essays on freedom and power, ed. G. Himmelfarb. Glencoe, IL: The Free Press. Elster, J., 1997. The market and the forum: Three varieties of political theory. In J. Bohman and W. Regh, eds. Deliberative democracy: Essays on reason and politics. Cambridge, Mass.: MIT Press. Ford, R.T, 1994. The boundaries of race: Political geography in legal analysis. Harvard Law Review, 107, 1843–1921. Foucault, M., 1979. [1975] Surveiller et punir: naissance de la prison. Paris: Gallimard. In A. Sheridan trans. Discipline and punish: The birth of the prison. New York: Vintage Books. Foucault, M., 1980 [1976] Histoire de la sexualité. Paris. In R. Hurley trans. The history of sexuality, Volume I: An introduction. New York: Vintage Books. Foucault, M., 1982. The subject and power,’ Critical Inquiry, 8, 777–795. Fraser, N., 1981., Foucault on modern power: Empirical insights and normative confusions. Praxis International, 1, 272–287. Gordon, C., ed., 1980. Power/knowledge: Selected interviews and other writings, 1972–1977. Trans. C. Gordon, L. Marshall, J. Mepham, and K. Soper. New York: Pantheon Books. Hay, C., 1997. Divided by a common language: Political theory and the concept of power. Politics, 17, 45–52. 20 C. Hayward and S. Lukes

Havel, Vaclav et al., 1985. The power of the powerless – citizens against the state in central-Eastern Europe, ed. John Keane. Amonk, NY: M.E. Sharpe. Hayward, C., 2000 De-facing power. Cambridge, UK: Cambridge University Press. Hayward, C., 2006. On power and responsibility. Political Studies Review, 4, 156–163. Jackson, K.T., 1987. Crabgrass frontier: The suburbanization of the United States NY: Oxford University Press. Lipsitz, G., 2006. The possessive investment in whiteness: How white people profit from identity politics. Philadelphia: Temple University Press. Lipton, E. et al., 2005. Breakdown marked path from hurricane to anarchy, New York Times, 11 September. Lukes, S., 1974. Power: A radical view. London: MacMillan. Lukes, S., 2005. Power: A radical view, second edition. Basingstoke: Palgrave Macmillan. Morriss, P., 2002. Power: A philosophical analysis, second edition. Manchester University Press. Plekhanov, G., 2003. [1898] The role of the individual in history. Honolulu, Hawaii: University Press of the Pacific. Polsby, N., 1980. Community power and political theory: A further look at problems of evidence and inference, second edition. New Haven: Yale University Press. Ricci, D., 1980. Receiving ideas in political analysis: The case of the community power studies, 1950–1970. Western Political Quarterly, 33, 451–475. Saulny, S., 2007. New Orleans hurt by acute rental shortage, New York Times, 12 December. Scott, J.C., 1985. Weapons of the weak: Everyday forms of peasant resistance. New Haven: Yale University Press. Self, R. and Sugrue, T., 2001. The power of place: Race, political economy, and identity in the postwar metropolis. In J.C. Agnew and R. Rozenzweig, eds. A companion to post-war America. Oxford: Blackwell, 20–43. Shapiro, I., 2003. The moral foundations of politics. New Haven: Yale University Press. Steinbeck, J., 1939. The grapes of wrath. New York: Penguin, 1992. Taylor, C., 1984. Foucault on freedom and truth. Political Theory, 12, 152–183. Tiebout, C., 1956. A pure theory of local expenditures. Journal of Political Economy, 64, 416–424. Walzer, M., 1983. The politics of Michel Foucault. Dissent, 30, 481–490. Young, I., 2004. Responsibility and global labor justice. Journal of Political Philosophy, 12, 365–388. Downloaded At: 02:36 30 January 2009